Constructing my Books Read in 2010 post

Constructing my previous post, Books Read in 2010, was far too difficult.  Still.

I keep a simple running list of books I read in VoodooPad, a personal wiki, on my laptop.  I record them by date started, author’s last name and title.  When I finish I add that date.  At some point I look them up in a library catalog—generally WorldCat nowadays—and bring them into Zotero and add them to a folder titled Books Read in 20xx.

In the past I have exported that folder from Zotero as a bibliography in HTML and pasted it into WordPress.  With some minor editing I got a decent bibliography including COinS data for every title.  But somewhere along the line over the last year or two things have gone wonky and some interaction between the COinS-formatted HTML from Zotero and WordPress have caused much of that data to be stripped out.  Last year was a real pain and seeing as this year my list was 20-some-odd percent longer I could not face all of that work simply to have much of it disappear no matter how much wrangling and struggling I did.

My next thought was that I would simply use the OpenBook plugin by John Miedema that I am using for book reviews [example post].  I was not looking forward to plugging one hundred or so ISBNs into its input form one at a time but it was in theory doable. [This was due more to how much work I had already done verifying ISBNs, “correcting” those in Zotero and pasting a copy of the ISBN into the text file created with the bibliography exported from Zotero than it was the effort to use the plugin.]

So I ran a little test trying a few “random” ISBNs from the list to see what the Open Library records looked like and/or if they had records for some of my less popular titles.  The results were horrible!  I estimated I would have to add records for at least 20 titles and fix records on 2 to 3 times that many.  I began slowly poking away at them over the course of a couple days—days when I should have been doing other things of course—and although my estimates were highly accurate I got it done.

At some point in my cataloging I noticed that Open Library had recently added a lists feature.  I thought perhaps I’ll just make a list there and point my blog readers to it; although that did strike me as rather dismal.  Of course, I noticed the list feature after I had added or re-cataloged somewhere around 30 books; which meant I had to look them all up again individually to add them to my new list.  ::sigh::

Then I discovered that you can export a list in either JSON, HTML or BibTex.  Sadly I know little to nothing about either JSON or BibTex so if they would have made my life easier—without a steep learning curve first—then I did myself a disfavor by using HTML.

Well, the HTML needed a lot of massaging to look decent once imported into WordPress.  As the native page exported by Open Library it looks OK, but WordPress treats those h3s, spans and divs much differently. [Technically not an export but a simplified page generated from your main list that you can save and/or copy from the source.]

I believe the titles are in the list in the order I entered them, or something close to that anyway.  Sadly, that order bears no relation to anything useful.  Thus, I had to cut and paste the whole list into the order I wanted.  Then I started playing with layout to see what would look decent enough in WordPress.  Once I figured it out I started changing the divs and h3s to spans and removing all the extraneous white space.  By hand.  TextEdit was of no use in the white space changing game.  As I was getting really tired of all the mousing, etc. involved I remembered that Dreamweaver might do a much better job with white space in find and replace.  With hope I fired up the long disused copy of DW and opened my file.  I highlighted a group of white space and a tag to change, hit ⌘-C to copy it, hit ⌘-F to open Find and Replace, saw that the white space was intact, put the cursor in the replace box, hit ⌘-V to paste the same in, deleted the white space I wanted removed, and hit OK.  It did what I wanted so I had it fix the rest of those and went on to the next bit needing fixed.  Thankfully Open Library had been consistent in how and where it added the white space.

After that it was rather simple to verify my data and do the odd minor correction here and there.  As for the ebooks, I pulled those out of my original list exported from Zotero and ginned them up in a text file with links to each book in feedbooks.  Yes, Open Library has ebooks but from what I could find not a single one from feedbooks.  I could have added them but I was in no mood to add another 18 books, and cataloging free ebooks that give absolutely no indication of which text they are was not something I intended to undertake.  Ebooks are great in many contexts!  Ebook metadata is in a despicable state! [That is a rant for another, and previous, occasion.]

Once I had the ebooks fully ginned up in the text file I cut and paste them into the blog post where they went in the list.  Then I wrote the text that went along with the list and waited for the end of the year a few days away.  On the 31st I made a few minor corrections to the list since I finished one of the books I had given up on and added another that I read on the 30th and 31st.  I also fixed the numbers/commentary regarding the other two books and added a bit more commentary.

Sadly, the only COinS data available is for the post itself and I doubt many of you are truly interested in adding my post to Zotero, Mendeley, or whatever.

If I had used OpenBook I could have had COinS.  But I got distracted by needing to fix so many records at Open Library and then by finding the Open Library list feature.  After spending so much time futzing and seeing what it would do for me I had given up on Open Library.  Honestly, I had no desire to copy and paste 100+ ISBNs into it one by one either.  Still, I wonder how well it would have handled the job? [John, if you are still reading, any idea how the plugin might handle 100+ titles using template 5, embedded? Certainly wouldn’t want to be making all those calls to OL live.]

None of this is meant to take away from the OpenBook plugin for which I greatly thank John Miedema!

It makes me sad that it is 2011 and this task is still so darn difficult.  Much progress has been made in the sharing and linking to book data on the web but it is still so crude.  Much of the assorted quasi-FRBRization going on in places like Open Library, WorldCat, goodreads, Library Thing, etc. actually seem to make it worse.  If one only cares about pointing at a title/work then things are somewhat better.  But I cared about editions long before I became a cataloger.  In most cases if someone takes a recommendation from me I could care less which edition of the work they read or listen to in the end.  But in some cases it does matter.  And for my own purposes I want to know which manifestation(s) of the work I engaged with.

Some day the future may arrive and making a list like this in which the titles will bring their own (accurate) metadata along with them will be easy. That day simply has to arrive. Soon.

Then again, I’m still waiting on the flying car I was promised almost 50 years ago.

10 thoughts on “Constructing my Books Read in 2010 post

  1. Hi Mark, I read almost every word in this post because I find the topic so interesting, if only because I have spent a lot of time thinking about these things through the development of OpenBook.

    I had thought about creating an OpenBook feature for calling many ISBNs at once. There is a question about display because people would want to insert text between books. (I also thought about an OpenBookShelf where you could collect your books over time, and then quickly drop them into your posts, sort of like the picture library.) But OpenBook development is on hold for the indefinite future. I am still supporting it, but not doing enhancements anytime soon (unless there was substantial interest from some quarter). I have too many other interests at present.

    With regard to Open Library, I will throw in one more wrinkle. I have noticed that updates to book cover images sometimes disappear, not immediately but a month or so later. Weird.

    Sorry, that doesn’t solve your problem. But I empathize.

  2. Oh, I don’t expect you to do anymore development, especially for such an (I’m assuming) odd use case. I know you have ceased development and I respect that.

    In fact, it is time for me to learn a bit more and step up to the plate. I just did a minor CSS edit to template 1, which is perhaps a baby step. Or not. 😉 The second thing will be to try a test post of perhaps 20 or so titles and see what happens.

    Anyway, I’m tired of griping about these issues and would like to get involved in some of these issues to help the world become more like I’d like to see it.

    First though, I need to decide if I am finishing this degree. If so then I have to write and defend my thesis before the end of the spring semester. *This* spring! Once I make a decision, or the deadline passes, I will be a bit freer to pursue some kind of involvement.

    I have a post in draft regarding some steps Sara and I are taking for self-learning. Hopefully our plan will pan out and some of the many things I need to learn about so I might be able to make a contribution will make it onto my active learning plate.

    Anyway, stay tuned.

  3. Code … or thesis? I would choose thesis, but that’s me. Then again, it’s funny that I chose to say this just today, “unless there was substantial interest from some quarter”. Maybe not quite what you’re looking for, but keep your ears on.

  4. Will do.

    And a thesis. Although I’m really struggling. Should have been done 2 years ago but life kind of got in the way. Having a real hard time getting back to the writing; the research is done.

  5. Yes, I was thinking of this earlier today, of how goodreads could also serve the same purpose, as could Open Books.

    I wonder what it is about my not wanting to do that? I guess I know I wouldn’t probably ever look at anyone’s list if all I got was a blog post/tweet/whatever that basically said, “Hey, here’s a link. Follow this to see my list over there.” That would sure be easier, to say the least.

  6. Since I started keeping track of what I read using GoodReads, I decided the easiest thing would be to set up a shelf by year. Now when I finish a book, make sure I add it to the current year’s shelf. Creating a list at the end of the year can be done with a link to the shelf or a widget if your blog software supports embedding them in posts.

  7. Hi, Anna. I’ve gone ahead and done something similar for this year. I may do just what you suggest at the end of the year.

    Perhaps I need to make myself go back and re-read these end-of-the-year plaints before starting to attempt this sort of silliness next year.

  8. Hi Mark, have you seen Google Refine? Having just watched the videos about it, it looks like you could have gotten your data cleaned up and automatically linked to Freebase very quickly.

Comments are closed.